Juvenile offenders involved in a system of care project: Predicting differential recidivism by youths' characteristics, family risk factors, emotional /behavioral indicators, early offending profiles, and service utilization

The overwhelming majority of system of care outcome research, to date, has examined the impact of this innovative service delivery model on all youths involved in such projects without concern for their heterogeneity. Although youths' differential characteristics and risk factors resulting in differential preliminary behavioral outcomes in system of care projects has been substantiated (Rosenblatt & Furlong, 1998), there is a paucity of research examining characteristics of and outcomes for specific populations of youths involved in system of care projects. Of particular interest in this study was the recidivism status of juvenile probation-referred youths receiving system of care interventions. The purpose of this study was to (a) examine juvenile offenders' sociodemographic, risk, emotional/behavioral, early offending, and service utilization profiles and (b) predict and classify these youths' recidivism status following one year of system of care involvement.;Data were collected from 178 juvenile probation-referred youths participating in Santa Barbara County's Multiagency System of Care (MISC). It is important to note that these juvenile offenders have had histories of multiple arrests at MISC intake, thus representing youths who have already entered a trajectory of reoffending. These youths were grouped by the number of rearrests after one year of MISC involvement ("0 rearrests" = Abstainers ; "1 rearrest" = Desisters; "2+ rearrests" = Persisters). Preliminary examination of these youths' profiles indicated that youths' risk factors (substance use and in-home placement), offense type (violation for property offense), and units of services received (assessment, case management, flexible services, and therapy) differed significantly across youths' post one year recidivism status. These variables along with other previously documented risk factors associated with recidivism were then included in a stepwise discriminant analysis. Results indicated that age at MISC referral, positive substance use history, in-home placement status, and units of case management services received all contributed to the reliable prediction of juvenile offenders' recidivism status post one year MISC entry with 53.3% of the youths correctly classified. Youths' age at first arrest, gender, offense severity, family risk factors, and number of prior arrests did not significantly discriminate between youths' recidivism status in the current study. Descriptions of youths' post one year recidivism profiles are presented to provide a meaningful context for these findings. Implications for juvenile justice, systems of care, and juvenile offenders are discussed, and topics for future research are suggested.